Error out if we are filtering a hostfile and encounter a node that is not in the resource-managed allocation, giving an error message identifying the file and the node. Don't filter managed allocations thru a default hostfile as this can lead to "hidden" errors.

Don't use dash-host info on managed allocations if we using soft locations

This commit was SVN r27245.

orte/mca/rmaps/base/rmaps_base_support_fns.c

@@ -87,7 +87,7 @@ int orte_rmaps_base_filter_nodes(orte_app_context_t *app,
         }
     }
     /* now filter the list through any -host specification */
-    if (NULL != app->dash_host) {
+    if (!orte_soft_locations && NULL != app->dash_host) {
         if (ORTE_SUCCESS != (rc = orte_util_filter_dash_host_nodes(nodes, app->dash_host, remove))) {
             ORTE_ERROR_LOG(rc);
             return rc;
@@ -407,38 +407,6 @@ int orte_rmaps_base_get_target_nodes(opal_list_t *allocated_nodes, orte_std_cntr
         return ORTE_ERR_SILENT;
     }
     
-    /* is there a default hostfile? */
-    if (NULL != orte_default_hostfile) {
-        OPAL_OUTPUT_VERBOSE((5, orte_rmaps_base.rmaps_output,
-                             "%s Filtering thru default hostfile",
-                             ORTE_NAME_PRINT(ORTE_PROC_MY_NAME)));
-
-        /* yes - filter the node list through the file, removing
-         * any nodes not in the file -or- excluded via ^
-         */
-        if (ORTE_SUCCESS != (rc = orte_util_filter_hostfile_nodes(allocated_nodes,
-                                                                  orte_default_hostfile,
-                                                                  true)) &&
-            ORTE_ERR_TAKE_NEXT_OPTION != rc) {
-            ORTE_ERROR_LOG(rc);
-            return rc;
-        }
-        OPAL_OUTPUT_VERBOSE((5, orte_rmaps_base.rmaps_output,
-                             "%s Resulted in %d nodes in list",
-                             ORTE_NAME_PRINT(ORTE_PROC_MY_NAME),
-                             (int)opal_list_get_size(allocated_nodes)));
-
-        /** check that anything is here */
-        if (0 == opal_list_get_size(allocated_nodes)) {
-            if (!silent) {
-                orte_show_help("help-orte-rmaps-base.txt",
-                               "orte-rmaps-base:no-available-resources",
-                               true);
-            }
-            return ORTE_ERR_SILENT;
-        }
-    }
- 
     /* filter the nodes thru any hostfile and dash-host options */
     OPAL_OUTPUT_VERBOSE((5, orte_rmaps_base.rmaps_output,
                          "%s Filtering thru apps",

orte/util/hostfile/help-hostfile.txt

@@ -10,6 +10,8 @@
 #                         University of Stuttgart.  All rights reserved.
 # Copyright (c) 2004-2005 The Regents of the University of California.
 #                         All rights reserved.
+# Copyright (c) 2012      Los Alamos National Security, LLC
+#                         All rights reserved.
 # $COPYRIGHT$
 # 
 # Additional copyrights may follow
@@ -109,3 +111,17 @@ Open RTE detected a bad parameter in the hostfile:
 The cores parameter is less than 0:
     cores=%d
 
+#
+[hostfile:extra-node-not-found]
+A hostfile was provided that contains at least one node not
+present in the allocation:
+
+  hostfile:  %s
+  node:      %s
+
+If you are operating in a resource-managed environment, then only
+nodes that are in the allocation can be used in the hostfile. You
+may find relative node syntax to be a useful alternative to
+specifying absolute node names see the orte_hosts man page for
+further information.
+

orte/util/hostfile/hostfile.c

@@ -565,7 +565,8 @@ int orte_util_filter_hostfile_nodes(opal_list_t *nodes,
     int num_empty, nodeidx;
     bool want_all_empty = false;
     opal_list_t keep;
-    
+    bool found;
+
     OPAL_OUTPUT_VERBOSE((1, orte_debug_output,
                         "%s hostfile: filtering nodes through hostfile %s",
                         ORTE_NAME_PRINT(ORTE_PROC_MY_NAME), hostfile));
@@ -717,6 +718,7 @@ int orte_util_filter_hostfile_nodes(opal_list_t *nodes,
              * search the provided list of nodes to see if this
              * one is found
              */
+            found = false;
             for (item1 = opal_list_get_first(nodes);
                  item1 != opal_list_get_end(nodes);
                  item1 = opal_list_get_next(item1)) {
@@ -747,9 +749,20 @@ int orte_util_filter_hostfile_nodes(opal_list_t *nodes,
                         /* mark as included */
                         node_from_list->mapped = true;
                     }
+                    found = true;
                     break;
                 }
             }
+            /* if the host in the newnode list wasn't found,
+             * then that is an error we need to report to the
+             * user and abort
+             */
+            if (!found) {
+                orte_show_help("help-hostfile.txt", "hostfile:extra-node-not-found",
+                               true, hostfile, node_from_file->name);
+                rc = ORTE_ERR_SILENT;
+                goto cleanup;
+            }
         }
         /* cleanup the newnode list */
         OBJ_RELEASE(item2);